APPLICATION OF ARTIFICIAL INTELLIGENCE IN PERSONALIZED HEALTHCARE

  1. INTRODUCTION

1.1 General

Artificial Intelligence (A.I.) is the advancement of technology, that can perform tasks that require human or our intelligence, like taking certain decisions, doing computation and complex operations, and learning new things.

As we know that artificial intelligence helps us in giving predictions with a high level of accuracy, also it provides help to perform complex problems which human takes days to solve.

The below diagram will help a lot to understand AI, ML, and DL.

Figure 1.1 Relationship between AI, ML, Neural Nets, Deep Learning

Example – Artificial Intelligence related robotic surgery

1.2 Impact of Artificial Intelligence in Health care

We can see the greatest impact of AI in the field of the health care sector. The latest report of PwC tells that artificial intelligence will contribute an additional fifteen point seven trillion to the world economy by 2030.

So there are generally two reasons that have made AI impactful in health care. The following two reasons are given below:

1. More amount of medical data – As we know that today is the time of digitalization. And most of our records are stored in electronic form. Same in the case of medical data. Today there are tons of medical data which are in the form of medical history, reports, and much more. Now by using the data implementing algorithms in AI-based on technologies i.e., deep learning, neural networks, and machine learning in hospitals, we are using applications of AI in personalized health care.

2. Introduction of complex algorithms – Machine Learning can not handle high-dimensional data such as medical or health care data. Also, we know that these data are very vast. But to analyze and process these thousands of data we need technologies like AI deep learning and neural networks. So this is how it became much easier for humans to understand.

This is how AI comes into the picture in the health care sector.

1.3 What is personalized healthcare?

It utilizes personalized planning to provide predictive, preventive, and personalized treatment. This approach is based on the person’s history, unique characteristics, and biology. Personalized health care can be implemented by using currently available technologies such as AI.

Today we see problems like some drugs work for some people and for others they cause some side effects or are not found effective. Also, the question arises in our mind such as why only some people develop the diseases like cancers, and others do not. It can be possible or true that genes and biological factors are the reason but solving such problems will be a great achievement for us.

Now in such cases mentioned above, medicine should deal with each patients’ illness in a personalized manner, on the basis of a person’s characteristics and biology. This whole process is known as precision medicine. It is used in personalized healthcare.

Now let’s discover how Artificial Intelligence playing important role in personalized health care and precision medicine.

2.0 Applications of AI in Personalized Healthcare

No doctor can have a grip on all 31 million medical papers out there but AI will assist them will help them analyze data which are in the form of medical history, reports, and much more.

2.1 Artificial Neuron Network (ANN)

The application of ANNs in personalized health care is the diagnosis, imaging, back pain, dementia, pathology, clinical diagnosis, prediction of cancer, speech recognition, prediction of length of stay, image analysis and interpretation, acute pulmonary embolism arrhythmias, or psychiatric disorders diseases. Some of the advantages of ANN as stated by are:

ANN stands for Artificial Neural networks which can learn linear and nonlinear models. It is found that the accuracy of linear and nonlinear models made by artificial neural networks can be measured statistically. Also, these models are flexible and can be easily updated. So they can be used in a dynamic environment like the health sector. Neural networks completely tolerate incomplete data and noise.

The major disadvantage of ANNs is that they are weak in providing insight into the structure or can say that black box algorithms. It cannot predict outside the range.

Fig 2.1 ANN Graphical Image

2.2 Machine Learning (ML)

Machine Learning comes under Artificial intelligence and it plays a vital role in health care. It can learn from data and take decisions without human intervention.

Machine learning is a part of health care right now whether we’re talking about the statistical models that doctors currently use as risk scores in the intensive care unit or we’re talking about more advanced high-capacity models that are being trained.

To understand what sort of risks are relevant for patients and what sort of treatments might be needed, machine learning and algorithms are a part of health care. Doctors get clinical data from practice and from knowledge by practice, they practice so if we could look at clinical records from a hospital from a clinic and see what sort of treatments are given, how patients are interacting with a healthcare team we could learn from that practice.

But then there’s also knowledge maybe we don’t just want to learn based on how doctors are practicing we may also want to look at the knowledge that’s been generated randomized controlled trials RCTs papers that are written textbooks right we could learn from both of those sources once we have that data we can train these simple statistical models or more advanced high-capacity models and then we can predict things important clinical events forecasting treatments that a patient might need those are really important for health care and progress so that’s what could happen.

One of the big sources of knowledge is a randomized controlled trial where you give one set of people treatment and see how well it works in that population however randomized controlled trials are very rare because they’re expensive so only 10 to 20 percent of the treatments that are given today are based on randomized controlled trials.

From the above paragraph, we can conclude that neither practice nor knowledge is perfect right now without any machine learning without technology.

So machine learning is used in precision medicine. To implement machine learning we require a lot of medical data for training. It takes time but it is totally worth it.

2.3 Natural Language Processing

The goal of artificial intelligence is to make human language understandable for computers. Now NLP is generally used in text translation and speech recognition tools. But NLP plays a vital role in analyzing unstructured medical notes to classify clinical documents and give various methods and provide useful insight like improving records of patients.

2.4 Analyzing Images or Data with AI

Radiologists can look at over 50-100 images per day but a trained computer can look over millions of images in hours it can even go through more data and it could not possible for a physician to cover this amount of data in his life.

AI revolution in CT or MRI scans and draw lines around tumors or pinpoints cells or designate ECG strips it’s a hard and monotonous task but it needs to be done for machines to utilize that data simply.

Data analyzers are the eyes of machine learning however there are really complicated tests that are hard to define think about spotting the tumor on a CT scan radiologists analyze the images take the patient’s medical records into consideration and have to keep so many things in mind before making a diagnosis for such tasks.

We have to turn to deep learning it’s a vastly more advanced method a deep learning algorithm can study raw images or perhaps it doesn’t even need images only the raw data coming straight out of the machine to analyze it.

While machine learning algorithms are blind without human help, deep learning algorithms only need to get trained at annotated data.  DL can handle unlabeled and unstructured data without human intervention based on artificial neural networks.

2.5 Artificial Intelligence in Genomics Sequencing

Genomics sequencing allows uncovering the genetic code which is about six billion letters and also to look for mistakes and defects in it that can be the drivers of disease like mutations in DNA which can cause cancer. Genome Sequencing is very helpful for treating diseases like cancer.

The disease can really be diagnosed in the DNA, genomic sequencing represented a molecular diagnosis. The first human genome took about 10 years to complete and now we’re doing it in about two days all the credit goes to this revolutionary technology called Artificial Intelligence which consists of machine learning and deep learning.

Now with the help of genomic sequencing, we can predict which drug works well for the patient and this is how we can use precision medicine. Cancer research is increasingly powered by data. Those data remain siloed across institutions.

2.6 AI in Personalized Treatment

The application of AI in health care is data management. Collecting, classifying, and tracing huge data sets of already available medical information.

Design treatment plans if able to analyze those data sets and combine them with attributes from a patient’s file to identify potential treatment plans, precision medicine classical medical practice put large groups of people in their focus and tries to develop clinical solutions drugs or treatment plans based on the needs of the statistically average person but with the ability to analyze vast amounts of medical information to achieve genome-sequencing health sensors and variables.

AI will most likely help healthcare move from the one-size-fits-all medical solutions towards targeted treatments per patient therapies and uniquely composed drugs. AI with revolutionized drug creation speaking of drugs pharmaceuticals – clinical trials take sometimes more than a decade and cost billions of dollars.

We can immensely speed up this process while making it a lot more cost-effective this will have an enormous effect on health care and how innovations reach everyday medicine just imagine how fast it could come up with a cure for the next pandemic if supercomputers and AI algorithms could help us in the process, health assistance and medication management.

Today, there are 200,000 leprosy patients diagnosed every year. If you catch it in the stage of a skin lesion, you can cure the person. We are teaching an AI algorithm to recognize leprosy in an image of a skin lesion. With AI, doctors can help patients help themselves to get to the right experts at the right time.

2.7 Deep Learning In Personalized Healthcare

Convolutional neural network which are cnns for short, those are used for analyzing images and the second type which is recurrence neural networks or rnns for shorts are used for analyzing what we call sequential data so sequential data, where the natural sequence within that data is important and that includes things like sequences of blood test results so we have blood tests at different points in times.

It also includes things like genetic sequences because each sequence comes after the previous point in the sequence. It also includes things like text because actually in a sentence the positioning is very important so having one word after another one is also very important and so the key benefit of these two approaches to deep learning in medicine is that it enables us to analyze new types of information images text sequential data and these kinds of data that we couldn’t really analyze very well before but now that we have these techniques it opens up a whole realm of possibilities within medicine.

3.1 Future Scope

The future scope of artificial intelligence in personalized health care is that we can use different trained algorithms which will save lives by analyzing thousands of previous patients’ data and provide detailed insight, personalized treatment, and medicine.

AI is today everywhere from google assistance to translation. Also when this technology came into health care, it completely changed the way of treatment and removed the human error.  In the future, we can expect major growth of AI in Genomics Sequencing. Rather it is currently being used in countries like the USA, and UK.

3.2 Conclusion

From this term paper, we can conclude that there is a lot of scope of artificial Intelligence in personalized healthcare. The only need is to implement it in our health sector. As we know that PwC tells that artificial intelligence will contribute an additional 15.7 trillion to the world economy by 2030. This data is clearly showing how fast we are implementing it.

More on Application of AI In Health Care

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC7580505/

https://www.nature.com/articles/s41746-019-0191-0

Share on: